WO2022070570A1

WO2022070570A1 - Image-processing device, image-processing method, and image-processing program

Info

Publication number: WO2022070570A1
Application number: PCT/JP2021/027309
Authority: WO
Inventors: 崇文小池
Original assignee: 富士フイルム株式会社
Priority date: 2020-09-30
Filing date: 2021-07-21
Publication date: 2022-04-07
Also published as: EP4223224A4; EP4223224A1; US20230215057A1; JPWO2022070570A1; JP7430814B2

Abstract

This image-processing device determines whether there is a tumor or a localized mass of a mammary gland in respective tumor candidate regions detected from a plurality of tomographic images representing a plurality of cross-sectional planes of a subject, selects a first tomographic image group from the plurality of tomographic images in a first region that has been determined to include a tumor, selects a second tomographic image group from the plurality of tomographic images in a second region that has been determined to include a localized mass of the mammary gland, selects a third tomographic image group from the plurality of tomographic images in a third region other than the first and second regions, and produces a composite two-dimensional image using the tomographic image groups selected for the first, second, and third regions.

Description

Image processing device, image processing method, and image processing program

The present disclosure relates to an image processing device, an image processing method, and an image processing program.

In recent years, in order to promote early detection of breast cancer, diagnostic imaging using a radiographic imaging device (called mammography) that photographs the breast has attracted attention. Further, in mammography, a tomographic image in which a desired tomographic plane is emphasized by moving a radiation source and irradiating the breast with radiation from a plurality of radiation source positions to take an image and reconstructing a plurality of projected images acquired thereby. Tomosynthesis imaging has been proposed to produce. In tomosynthesis imaging, multiple source positions can be moved by moving the radiation source parallel to the radiation detector or in a circular or elliptical arc, depending on the characteristics of the imaging device and the required tomographic image. Acquire multiple projection images by photographing the breast in. Then, a plurality of acquired projected images are reconstructed to generate a tomographic image by using a back projection method such as a simple back projection method or a filter back projection method, or a sequential reconstruction method.

By generating such a tomographic image on a plurality of tomographic planes in the breast, it is possible to separate the overlapping structures in the breast in the depth direction in which the tomographic planes are lined up. For this reason, abnormalities such as lesions that were difficult to detect in a two-dimensional image (hereinafter referred to as "simple two-dimensional image") acquired by conventional simple imaging that irradiates the subject with radiation from a predetermined direction. It becomes possible to discover the site.

In addition, multiple tomographic images with different distances (positions in the height direction) from the detection surface of the radiation detector to the radiation source side acquired by tomosynthesis imaging can be combined with the addition method, averaging method, maximum value projection method, or minimum. A technique for generating a pseudo two-dimensional image (hereinafter referred to as "composite two-dimensional image") corresponding to a simple two-dimensional image by synthesizing by a value projection method or the like is known (Japanese Patent Laid-Open No. 2014-128716). See publication).

In addition, the mass lesion and the normal breast structure are detected from the tomographic image, the mass lesion is emphasized more than the normal breast structure and expressed on the synthetic two-dimensional image, and the normal breast structure is synthesized so as to represent the simple two-dimensional image 2. A technique for expressing on a two-dimensional image is known (see Japanese Patent Publication No. 2020-512129).

However, in conventional synthetic two-dimensional images produced by the method described in JP-A-2020-512129, for example, a local mammary gland mass that looks like a mass lesion in a simple two-dimensional image is still a mass lesion. It looks like. That is, the conventional synthetic two-dimensional image may not have the same diagnostic ability as the tomographic image. For this reason, a diagnostician such as a doctor will interpret not only a synthetic two-dimensional image but also a large number of tomographic images, which increases the load of interpretation.

The present disclosure has been made in view of the above circumstances, and provides an image processing device, an image processing method, and an image processing program capable of generating a synthetic two-dimensional image having a diagnostic ability equivalent to that of a tomographic image. The purpose is.

The image processing apparatus of the present disclosure is an image processing apparatus including at least one processor, and the processor detects a mass candidate region from a plurality of tomographic images representing a plurality of tomographic planes of a subject, and each of the detected mass candidate regions. It is determined whether the image is a mass or a local mass of the mammary gland, and in the first region determined to be a mass, the first tomographic image group is selected from a plurality of tomographic images, and the local mammary gland is selected. A second tomographic image group is selected from a plurality of tomographic images in the second region determined to be a mass of, and a third tomographic image other than the first region and the second region is selected from a plurality of tomographic images. A third tomographic image group is selected, and a synthetic two-dimensional image is generated using the tomographic image group selected for each of the first region, the second region, and the third region.

In the image processing apparatus of the present disclosure, the first tomographic image group may be a tomographic image group in which a tumor candidate region determined to be a tumor is detected among a plurality of tomographic images.

Further, in the image processing apparatus of the present disclosure, the second tomographic image group is a tomographic image group other than the tomographic image in which the mass candidate region determined to be a local mass of the mammary gland among a plurality of tomographic images is detected. It may be.

Further, in the image processing apparatus of the present disclosure, the second tomographic image group is a tomographic image in which a mass candidate region determined to be a local mass of the mammary gland among a plurality of tomographic images is detected, and the tomographic image thereof. It may be a tomographic image of a layer adjacent to.

Further, in the image processing apparatus of the present disclosure, the second tomographic image group is the same tomographic image group as the third tomographic image group, and the processor adjusts the density of the second region to a density close to that of the third region. A controlled composite two-dimensional image may be generated.

Further, in the image processing apparatus of the present disclosure, the third tomographic image group has the average value of the pixel value of the attention pixel of all the plurality of tomographic images and the plurality of tomographic images and the pixel value of the attention pixel of all the plurality of tomographic images. A tomographic image group in which the absolute value of the difference between the image and the above is equal to or higher than a preset threshold value, a tomographic image group having a preset number of tomographic images in descending order of the dispersion value of the pixel value of the region of interest including the pixel of interest among a plurality of tomographic images. It may be a tomographic image group having pixels whose edges are detected by edge detection processing among a plurality of tomographic images.

Further, the image processing apparatus of the present disclosure controls the processor to display the generated synthetic two-dimensional image, and determines whether the synthetic two-dimensional image is a mass or a local mass of the mammary gland on the synthetic two-dimensional image. Control to display the result may be performed.

Further, the image processing method of the present disclosure detects a mass candidate region from a plurality of tomographic images representing a plurality of tomographic planes of a subject, and each of the detected mass candidate regions is a mass or a local mass of the mammary gland. In the first region determined to be a mass, the first tomographic image group was selected from a plurality of tomographic images, and in the second region determined to be a local mass of the mammary gland, a plurality of tomographic images were selected. A second tomographic image group is selected from the tomographic images of the above, and a third tomographic image group is selected from a plurality of tomographic images in a third region other than the first region and the second region, and the first region is selected. , The process of generating a composite two-dimensional image using the tomographic image group selected for each of the second region and the third region is executed by the processor included in the image processing apparatus.

Further, the image processing program of the present disclosure detects a mass candidate region from a plurality of tomographic images representing a plurality of tomographic planes of a subject, and each of the detected mass candidate regions is a mass or a local mass of the mammary gland. In the first region determined to be a mass, the first tomographic image group was selected from a plurality of tomographic images, and in the second region determined to be a local mass of the mammary gland, a plurality of tomographic images were selected. A second tomographic image group is selected from the tomographic images of the above, and a third tomographic image group is selected from a plurality of tomographic images in a third region other than the first region and the second region, and the first region is selected. This is for causing a processor included in the image processing apparatus to perform a process of generating a composite two-dimensional image using a tomographic image group selected for each of the second region and the third region.

According to the present disclosure, it is possible to generate a synthetic two-dimensional image having a diagnostic ability equivalent to that of a tomographic image.

It is a schematic block diagram of a radiation imaging system. It is a figure which looked at the radiation imaging apparatus in the direction of arrow A of FIG. It is a block diagram which shows an example of the hardware composition of an image processing apparatus. It is a block diagram which shows an example of the functional structure of an image processing apparatus. It is a figure for demonstrating the acquisition process of a projection image. It is a figure for demonstrating the generation process of a tomographic image. It is a figure which shows an example of a plurality of tomographic images. It is a figure for demonstrating the generation process of a synthetic 2D image. It is a figure which shows an example of a synthetic 2D image. It is a flowchart which shows an example of the synthetic 2D image generation processing. It is a figure which shows an example of a synthetic 2D image. It is a figure which shows an example of a synthetic 2D image. It is a flowchart which shows another example of the synthetic 2D image generation processing. It is a figure which shows an example of the display state of the determination result. It is a figure which shows an example of the display state of the determination result. It is a figure which shows an example of the display state of the determination result.

Hereinafter, an example of a form for implementing the technique of the present disclosure will be described in detail with reference to the drawings.

First, the configuration of the radiographic imaging system 100 according to the present embodiment will be described with reference to FIGS. 1 and 2. As shown in FIGS. 1 and 2, the radiographic imaging system 100 photographs breast M, which is a subject, from a plurality of radiation source positions in order to perform tomosynthesis imaging of the breast and generate a tomographic image, and a plurality of breasts M are photographed. It is for acquiring a radiographic image, that is, a plurality of projected images. The radiographic imaging system 100 includes a mammography imaging apparatus 1, a console 2, an image storage system 3, and an image processing apparatus 4.

The mammography photographing apparatus 1 includes an arm portion 12 connected to a base (not shown) by a rotating shaft 11. An imaging table 13 is attached to one end of the arm portion 12, and a radiation irradiation unit 14 is attached to the other end so as to face the photographing table 13. The arm portion 12 is configured to be able to rotate only the end portion to which the radiation irradiation unit 14 is attached, whereby it is possible to fix the photographing table 13 and rotate only the radiation irradiation unit 14. It has become.

A radiation detector 15 such as a flat panel detector is provided inside the photographing table 13. The radiation detector 15 has a radiation detection surface 15A. Further, inside the photographing table 13, a charge amplifier that converts the charge signal read from the radiation detector 15 into a voltage signal, a correlated double sampling circuit that samples the voltage signal output from the charge amplifier, and an analog A circuit board or the like provided with an AD (Analog-to-Digital) conversion unit or the like that converts a voltage signal into a digital signal is also installed.

The radiation source 16 is housed inside the radiation irradiation unit 14. The radiation source 16 emits X-rays as radiation, and the timing of irradiating radiation from the radiation source 16 and the radiation generation conditions in the radiation source 16, that is, the selection of the material of the target and the filter, the tube voltage, the irradiation time, and the like are determined. It is controlled by the console 2.

Further, in the arm portion 12, a compression plate 17 arranged above the imaging table 13 to press and press the breast M, a support portion 18 for supporting the compression plate 17, and a support portion 18 are shown in FIGS. 1 and 2. A moving mechanism 19 for moving in the vertical direction is provided. The distance between the compression plate 17 and the imaging table 13, that is, the compression breast thickness is input to the console 2.

The console 2 displays the shooting order and various information acquired from an unillustrated RIS (Radiology Information System) or the like via a network such as a wireless communication LAN (Local Area Network), and instructions given directly by an engineer or the like. It has a function of controlling the mammography photographing apparatus 1 by using the device. Specifically, the console 2 acquires a plurality of projected images as described later by causing the mammography imaging device 1 to perform tomosynthesis imaging of the breast M, reconstructs the plurality of projected images, and performs a plurality of tomographic images. To generate. As an example, in this embodiment, the server computer is used as the console 2.

The image storage system 3 is a system that stores image data such as a radiographic image and a tomographic image taken by the mammography photographing apparatus 1. The image storage system 3 extracts the image data in response to the request from the console 2, the image processing device 4, and the like from the stored image data, and transmits the image data to the requesting device. Specific examples of the image storage system 3 include PACS (Picture Archiving and Communication Systems).

Next, the hardware configuration of the image processing device 4 according to the present embodiment will be described with reference to FIG. As shown in FIG. 3, the image processing device 4 includes a CPU (Central Processing Unit) 20, a memory 21 as a temporary storage area, and a non-volatile storage unit 22. Further, the image processing device 4 includes a display 23 such as a liquid crystal display, an input device 24 such as a keyboard and a mouse, and a network I / F (InterFace) 25 connected to the network. The CPU 20, the memory 21, the storage unit 22, the display 23, the input device 24, and the network I / F 25 are connected to the bus 27.

The storage unit 22 is realized by an HDD (Hard Disk Drive), an SSD (Solid State Drive), a flash memory, or the like. The image processing program 30 is stored in the storage unit 22 as a storage medium. The CPU 20 reads the image processing program 30 from the storage unit 22, expands it into the memory 21, and executes the expanded image processing program 30.

Next, with reference to FIG. 4, the functional configuration of the image processing apparatus 4 according to the present embodiment will be described. As shown in FIG. 4, the image processing device 4 includes an acquisition unit 40, a detection unit 42, a determination unit 44, a selection unit 46, a composition unit 48, and a display control unit 50. When the CPU 20 executes the image processing program 30, it functions as an acquisition unit 40, a detection unit 42, a determination unit 44, a selection unit 46, a composition unit 48, and a display control unit 50.

The acquisition unit 40 acquires a plurality of tomographic images generated by the console 2 causing the mammography imaging apparatus 1 to perform tomosynthesis imaging. The acquisition unit 40 acquires a plurality of tomographic images from the console 2 or the image storage system 3 via the network I / F25.

Here, tomosynthesis imaging and tomographic image generation processing on the console 2 will be described. The console 2 moves the radiation source 16 by rotating the arm portion 12 around the rotation axis 11 when performing tomosynthesis imaging for generating a tomographic image. Further, the console 2 irradiates the breast M, which is a subject, with radiation under predetermined imaging conditions for tomosynthesis imaging at a plurality of radiation source positions due to the movement of the radiation source 16. Further, the console 2 has a plurality of projected images Gi (i = 1 to n, n at the radiation source positions) at the plurality of radiation source positions obtained by detecting the radiation transmitted through the breast M by the radiation detector 15. It is a number, for example, n = 15) is acquired.

As shown in FIG. 5, the radiation source 16 is moved to each source position of Si (i = 1 to n), and the radiation source 16 is driven at each source position to irradiate the breast M with radiation. When the radiation detector 15 detects the radiation transmitted through the breast M, the projected images G1, G2, ..., Gn are acquired corresponding to the source positions S1 to Sn. At each source position S1 to Sn, the same dose of radiation is applied to the breast M.

Note that, in FIG. 5, the radiation source position Sc is the radiation source position where the optical axis X0 of the radiation emitted from the radiation source 16 is orthogonal to the detection surface 15A of the radiation detector 15. Hereinafter, the radiation source position Sc is referred to as a reference radiation source position Sc.

The console 2 generates a plurality of tomographic images that emphasize the desired tomographic plane of the breast M by reconstructing the plurality of projected images Gi. Specifically, the console 2 reconstructs a plurality of projected images Gi by using a well-known back projection method such as a simple back projection method or a filter back projection method. As a result, as shown in FIG. 6, the console 2 generates a plurality of tomographic images Dj (j = 1 to m) representing each of the plurality of tomographic planes of the breast M. At this time, a three-dimensional coordinate position in the three-dimensional space including the breast M is set, and the pixel values of the corresponding pixels of the plurality of projected images Gi are reconstructed with respect to the set three-dimensional coordinate position. The pixel value at that coordinate position is calculated.

The console 2 transfers the generated tomographic image Dj to the image processing device 4 or transfers it to the image storage system 3.

As shown in FIG. 7, the detection unit 42 detects a tumor candidate region from a plurality of tomographic images Dj acquired by the acquisition unit 40. In FIG. 7, for the sake of simplicity, the case where the number of tomographic images Dj is 5 (that is, j = 1 to 5) is illustrated. Further, in the example of FIG. 7, the tomographic image D2 includes the tumor candidate region K21, the tomographic image D3 includes the tumor candidate regions K31 and K32, and the tomographic image D4 includes the tumor candidate region K41. Further, in the example of FIG. 7, the tomographic image D4 includes the mammary gland N41. Further, it is assumed that the tumor candidate region K21, the tumor candidate region K31, and the tumor candidate region K41 have substantially the same position of the center of gravity in each tomographic image Dj. Further, in the example of FIG. 7, the tumor candidate region K21, the tumor candidate region K31, and the tumor candidate region K41 are tumors, and the tumor candidate region K32 is a local mass of the mammary gland.

The detection unit 42 detects a mass candidate region from a plurality of tomographic images Dj using a known computer-aided diagnosis (CAD: Computer Aided Diagnosis) mass detection algorithm. In the algorithm for detecting a mass by CAD, a probability (likelihood) indicating that a pixel in the tomographic image Dj is a mass candidate region is derived, and a pixel whose probability is equal to or higher than a predetermined threshold is used as a mass candidate region. Detected. Since it is difficult to distinguish between a mass and a local mass of the mammary gland with only one tomographic image Dj, the local mass of the mammary gland is also detected as a tumor candidate region.

The detection of the tumor candidate region is not limited to the one using CAD. The tumor candidate region is detected from the tomographic image Dj by filtering processing by a filter for detecting the tumor candidate region, or by a detection model in which machine learning is performed by deep learning or the like to detect the tumor candidate region. May be good.

The determination unit 44 determines whether each of the tumor candidate regions detected by the detection unit 42 is a tumor or a local mammary gland mass. Since the mass is a three-dimensional structure in which the inside is filled with cancer cells, which is generated by the proliferation of cancer cells, the tomographic image Dj appears over a plurality of continuous tomographic images Dj. On the other hand, the local mammary gland mass has a structure in which thinly stretched mammary glands are overlapped, so that it looks like a mass on one tomographic image Dj, but it looks like a normal mammary gland on the tomographic image Dj of the adjacent layer. It is reflected in.

Therefore, the determination unit 44 according to the present embodiment determines that the tumor candidate region which is reflected in a plurality of continuous tomographic images Dj and whose center of gravity is almost the same in each tomographic image Dj is a tumor. .. The position where the positions of the centers of gravity are substantially the same here means that, for example, the distance between the positions of the centers of gravity of the tumors reflected in each of the plurality of tomographic images Dj is equal to or less than a predetermined threshold. As the threshold value in this case, for example, a preset value can be applied as an upper limit value of the distance between the positions of the centers of gravity of the tumors reflected in a plurality of continuous tomographic images Dj. Further, the determination unit 44 determines that the tumor candidate region, which is not shown in the tomographic image Dj of the adjacent layer but is shown only in one tomographic image Dj, is a local mammary gland mass.

In the example of FIG. 7, the determination unit 44 is shown in the tomographic images D2 to D4 in which the tumor candidate regions K21, K31, and K41 are continuous, and the positions of the centers of gravity in the respective tomographic images D2 to D4 are almost the same. Tumor candidate regions K21, K31, and K41 are determined to be tumors. Further, in the example of FIG. 7, the determination unit 44 does not show the tumor candidate region K32 in the tomographic image D3 at almost the same position as the tomographic images D2 and D4 in the layer adjacent to the tomographic image D3. Therefore, the tumor candidate region K32 is determined to be a local mammary gland mass.

The selection unit 46 selects a first tomographic image group from a plurality of tomographic images Dj in the first region determined to be a tumor by the determination unit 44. In this selection, the selection unit 46 selects the first tomographic image group according to the first selection rule. The first selection rule according to the present embodiment is a rule to select a tomographic image group in which a tumor candidate region determined to be a tumor is detected by the determination unit 44 from a plurality of tomographic images Dj. In the example of FIG. 7, the selection unit 46 selects the tomographic images D2 to D4 in which the tumor candidate regions K21, K31, and K41 are detected as the first tomographic image group.

The selection unit 46 selects a second tomographic image group from a plurality of tomographic images Dj in the second region determined by the determination unit 44 to be a local mammary gland mass. In this selection, the selection unit 46 selects the second tomographic image group according to a second selection rule different from the first selection rule. The second selection rule according to the present embodiment is a tomographic image group other than the tomographic image in which the tumor candidate region determined to be a local mammary gland mass by the determination unit 44 is detected among the plurality of tomographic images Dj. Is the rule to select. In the example of FIG. 7, the selection unit 46 selects tomographic images D1, D2, D4, and D5 other than the tomographic image D3 in which the tumor candidate region K32 is detected as the second tomographic image group.

The selection unit 46 selects a third tomographic image group from a plurality of tomographic images Dj in a third region other than the first region and the second region. In this selection, the selection unit 46 selects a third tomographic image group according to a third selection rule different from the first selection rule and the second selection rule. The third selection rule according to the present embodiment is a rule that all the plurality of tomographic images Dj are selected. In the example of FIG. 7, the selection unit 46 selects tomographic images D1 to D5 as the third tomographic image group.

The synthesis unit 48 generates a composite two-dimensional image using the tomographic image group selected by the selection unit 46 for each of the first region, the second region, and the third region. The synthetic two-dimensional image is a pseudo two-dimensional image corresponding to a simple two-dimensional image taken by irradiating the breast M with radiation from the reference radiation source position Sc. In the present embodiment, as shown in FIG. 8, the synthesis unit 48 has a state in which a plurality of tomographic images Dj are stacked, and is shown in the viewpoint direction from the reference source position Sc toward the radiation detector 15, that is, in FIG. A composite two-dimensional image is generated by synthesizing the pixel values of the corresponding pixels in each tomographic image Dj along the optical axis X0. Hereinafter, a specific example of the generation process of the composite two-dimensional image will be described.

FIG. 9 shows an example of the composite two-dimensional image CG0. As shown in FIG. 9, for the first region A1, tumor candidate regions K21, K31, and K41 determined to be tumors in the tomographic images D2 to D4 are synthesized. At the time of this synthesis, the synthesis unit 48 uses the added average value of the pixel values as the pixel value of the composite two-dimensional image CG0 for the pixels at the corresponding pixel positions of the plurality of tumor candidate regions. Further, at the time of this synthesis, the synthesis unit 48 uses the pixel value of the pixel as the pixel value of the composite two-dimensional image CG0 for the pixel at the pixel position included in only one of the plurality of tumor candidate regions. In addition, for example, in the synthesis unit 48, among the plurality of mass candidate regions, the largest region, the smallest region, or the average region of each region is set as the first region A1, and the first region of the selected tomographic image group is selected. The added average value of the pixel values of each pixel in the region A1 may be used as the pixel value of the composite two-dimensional image CG0.

Further, as shown in FIG. 9, the second region A2 corresponds to the tumor candidate region K32 of the tomographic image D3 determined to be a local mammary gland mass in the tomographic images D1, D2, D4, and D5. Areas to be combined. At the time of this composition, the composition unit 48 sets the added average value of the pixel values as the pixel value of the composite two-dimensional image CG0 for each pixel included in the second region A2 of each of the tomographic images D1, D2, D4, and D5. ..

Further, as shown in FIG. 9, for the third region A3, the corresponding regions of the tomographic images D1 to D5 are synthesized. At the time of this synthesis, the synthesis unit 48 sets the added average value of the pixel values as the pixel value of the composite two-dimensional image CG0 for each pixel included in the third region A3 of each of the tomographic images D1 to D5.

The display control unit 50 controls to display the composite two-dimensional image CG0 generated by the composite unit 48 on the display 23.

Next, with reference to FIG. 10, the operation of the image processing apparatus 4 according to the present embodiment will be described. When the CPU 20 executes the image processing program 30, the synthetic two-dimensional image generation process shown in FIG. 10 is executed. The composite two-dimensional image generation process shown in FIG. 10 is executed, for example, when an instruction to start execution is input by the user via the input device 24.

In step S10 of FIG. 10, the acquisition unit 40 acquires a plurality of tomographic images Dj generated by the console 2 causing the mammography imaging apparatus 1 to perform tomosynthesis imaging. In step S12, as described above, the detection unit 42 detects the tumor candidate region from the plurality of tomographic images Dj acquired in step S10.

In step S14, as described above, the determination unit 44 determines whether each of the tumor candidate regions detected in step S12 is a tumor or a local mammary gland mass. In step S16, as described above, the selection unit 46 selects the first tomographic image group from the plurality of tomographic images Dj in the first region determined to be a tumor in step S14 according to the first selection rule. do. Further, as described above, the selection unit 46 has a plurality of tomographic images Dj to a second tomographic image in the second region determined to be a local mammary gland mass in step S14 according to the second selection rule. Select a group. Further, as described above, the selection unit 46 selects the third tomographic image group from the plurality of tomographic images Dj in the third region other than the first region and the second region according to the third selection rule. .. Then, as described above, the synthesis unit 48 generates a composite two-dimensional image CG0 for each of the first region, the second region, and the third region using the tomographic image group selected by the selection unit 46. do.

In step S18, the display control unit 50 controls to display the composite two-dimensional image CG0 generated in step S16 on the display 23. When the process of step S18 is completed, the composite two-dimensional image generation process is completed.

As described above, in the synthetic two-dimensional image generation processing of this example, the second tomographic image group used for the generation of the second region A2 of the synthetic two-dimensional image CG0 is the local mammary gland among the plurality of tomographic images Dj. It is a tomographic image group (tomographic images D1, D2, D4, and D5) other than the tomographic image D3 in which the mass candidate region K32 determined to be a mass is detected. Therefore, the local mass of the mammary gland that looks like a mass in the simple 2D image is not represented by the synthetic 2D image CG0 shown in FIG. Therefore, in the synthetic two-dimensional image CG0 shown in FIG. 9, it is suppressed that the local mass of the mammary gland is erroneously recognized as a mass.

As described above, according to the present embodiment, it is possible to generate a synthetic two-dimensional image having a diagnostic ability equivalent to that of a tomographic image. As a result, a diagnostician such as a doctor need only interpret one composite two-dimensional image, so that the load of interpretation is reduced. In addition, by deleting the tomographic image that is no longer needed for image interpretation from the storage device, the capacity of the storage device can be saved.

In the above embodiment, as the second tomographic image group, a tomographic image group other than the tomographic image in which the mass candidate region determined to be a local mammary gland mass among the plurality of tomographic images Dj is detected is applied. However, the case is not limited to this. For example, as a second tomographic image group, a tomographic image in which a mass candidate region determined to be a local mass of the mammary gland among a plurality of tomographic images Dj is detected, and a tomographic image of a layer adjacent to the tomographic image. And may be applied as a form. In this case, as an example, as in the synthetic two-dimensional image CG0 shown in FIG. 11, in the second region A2, the tumor candidate region K32 of the tomographic image D3 determined to be a local mammary gland mass and the tomographic image. The mammary gland N41 of the tomographic image D4 of the layer adjacent to D3 is synthesized. Therefore, in the synthetic two-dimensional image CG0 shown in FIG. 11, the tumor candidate region K32 and the mammary gland N41 are synthesized, so that the tumor candidate region K32 is clearly expressed as a local mass of the mammary gland. Therefore, even in the synthetic two-dimensional image CG0 shown in FIG. 11, it is suppressed that the local mass of the mammary gland is erroneously recognized as a tumor.

Further, for example, as the second tomographic image group, the same tomographic image group as the third tomographic image group may be applied. In this case, the second selection rule and the third selection rule are the same rule. Further, in this case, the synthesis unit 48 generates a synthetic two-dimensional image CG0 in which the density of the second region A2 in the synthetic two-dimensional image CG0 is controlled to be close to the density of the third region A3. In this case, as shown in FIG. 12, as an example, the density of the pixels in the second region A2 is closer to the third region A3 than the first region A1, so that the pixels are clearly defined as the tumor region. Can be distinguished. Therefore, even in the synthetic two-dimensional image CG0 shown in FIG. 12, it is suppressed that the local mass of the mammary gland is erroneously recognized as a tumor.

Further, as shown in the flowchart shown in FIG. 10, not only the composite two-dimensional image CG0 is displayed, but also the display control is added to the step S18 for displaying the composite two-dimensional image CG0 as shown in the flowchart shown in FIG. The unit 50 may execute step S20 for displaying the determination result by the determination unit 44. In step S20, the display control unit 50 controls to display the determination result of whether the tumor is a mass or a local mammary gland mass by the determination unit 44 on the synthetic two-dimensional image CG0. FIG. 14 shows an example of the display state of the determination result. As shown in FIG. 14, the display control unit 50 controls to display the character string representing the determination result by the determination unit 44 in the vicinity of the tumor candidate region to be determined. In the example of FIG. 14, the character string "mammary gland" is displayed at the lower part of the tumor candidate area determined to be a local mass of the mammary gland, and the "tumor" is displayed at the lower part of the tumor candidate area determined to be a mass. "Is displayed.

Further, as shown in FIG. 15, the display control unit 50 may control to display the determination result by the determination unit 44 by changing the color of the border of the tumor candidate region to be determined. In the example of FIG. 15, the border of the tumor candidate region determined to be a local mammary gland mass is displayed as a broken line, and the border of the tumor candidate region determined to be a mass is displayed as a dotted chain line. However, this means that the color of the border is different.

Further, as shown in FIG. 16, the display control unit 50 may control to display the mark indicating the determination result by the determination unit 44 in the vicinity of the tumor candidate region to be determined. In the example of FIG. 16, a circle mark is displayed at the bottom of the tumor candidate area determined to be a local mammary gland mass, and a triangular mark is displayed at the bottom of the tumor candidate area determined to be a mass. ing.

Further, in the above embodiment, the case where all of the plurality of tomographic images Dj are applied as the third tomographic image group has been described, but the present invention is not limited to this. For example, as a third tomographic image group, an absolute value of the difference between the pixel value of the pixel of interest among the plurality of tomographic images Dj and the average value of the pixel values of all the pixels of interest of the plurality of tomographic images Dj is set in advance. The above tomographic image group may be applied. Further, for example, a predetermined number of tomographic image groups may be applied in descending order of the dispersion value of the pixel values in the region of interest including the pixel of interest among the plurality of tomographic images Dj. Further, for example, as the third tomographic image group, a tomographic image group having pixels whose edges are detected by the edge detection process among a plurality of tomographic images Dj may be applied.

Further, in the above embodiment, the case where the added average value of the pixel values of the corresponding pixels of the selected tomographic image group is applied as the pixel value of the pixel of the composite two-dimensional image CG0 has been described, but the present invention is not limited to this. As the pixel value of the pixel of the composite two-dimensional image CG0, the median value, the maximum value, or the minimum value of the pixel value of the corresponding pixel of the selected tomographic image group may be applied.

Further, in the above embodiment, for example, the hardware of the processing unit (processing unit) that executes various processes such as the acquisition unit 40, the detection unit 42, the determination unit 44, the selection unit 46, the synthesis unit 48, and the display control unit 50. As the structure, various processors shown below can be used. As described above, the various processors include a CPU, which is a general-purpose processor that executes software (program) and functions as various processing units, and a circuit after manufacturing an FPGA (Field Programmable Gate Array) or the like. Dedicated electricity, which is a processor with a circuit configuration specially designed to execute specific processing such as programmable logic device (PLD), ASIC (Application Specific Integrated Circuit), which is a processor whose configuration can be changed. Circuits etc. are included.

One processing unit may be composed of one of these various processors, or a combination of two or more processors of the same type or different types (for example, a combination of a plurality of FPGAs or a combination of a CPU and an FPGA). It may be composed of a combination). Further, a plurality of processing units may be configured by one processor.

As an example of configuring a plurality of processing units with one processor, first, one processor is configured by a combination of one or more CPUs and software, as represented by a computer such as a client and a server. There is a form in which the processor functions as a plurality of processing units. Second, as typified by System on Chip (SoC), there is a form that uses a processor that realizes the functions of the entire system including multiple processing units with one IC (Integrated Circuit) chip. be. As described above, the various processing units are configured by using one or more of the above-mentioned various processors as a hardware-like structure.

Further, as the hardware structure of these various processors, more specifically, an electric circuit (circuitry) in which circuit elements such as semiconductor elements are combined can be used.

Further, in the above embodiment, the embodiment in which the image processing program 30 is stored (installed) in the storage unit 22 in advance has been described, but the present invention is not limited to this. The image processing program 30 is provided in a form recorded on a recording medium such as a CD-ROM (Compact Disc Read Only Memory), a DVD-ROM (Digital Versatile Disc Read Only Memory), and a USB (Universal Serial Bus) memory. May be good. Further, the image processing program 30 may be downloaded from an external device via a network.

The disclosure of Japanese Patent Application No. 2020-166473 filed on September 30, 2020 is incorporated herein by reference in its entirety. In addition, all documents, patent applications, and technical standards described herein are to the same extent as if it were specifically and individually stated that the individual documents, patent applications, and technical standards were incorporated by reference. Is incorporated herein by reference.

Claims

An image processing device equipped with at least one processor,
The processor
Tumor candidate areas are detected from multiple tomographic images representing multiple tomographic planes of the subject.
For each of the detected tumor candidate areas, it is determined whether it is a tumor or a local mass of the mammary gland.
In the first region determined to be a tumor, the first tomographic image group is selected from the plurality of tomographic images.
In the second region determined to be a local mammary gland mass, a second tomographic image group is selected from the plurality of tomographic images.
In the third region other than the first region and the second region, a third tomographic image group is selected from the plurality of tomographic images.
An image processing device that generates a synthetic two-dimensional image using a tomographic image group selected for each of the first region, the second region, and the third region.
The image processing apparatus according to claim 1, wherein the first tomographic image group is a tomographic image group in which the tumor candidate region determined to be a tumor is detected among the plurality of tomographic images.
The second tomographic image group is a tomographic image group other than the tomographic image in which the mass candidate region determined to be a local mass of the mammary gland among the plurality of tomographic images is detected. Item 2. The image processing apparatus according to Item 2.
The second tomographic image group includes a tomographic image in which the mass candidate region determined to be a local mass of the mammary gland among the plurality of tomographic images is detected, and a tomographic image of a layer adjacent to the tomographic image. The image processing apparatus according to claim 1 or claim 2.
The second tomographic image group is the same tomographic image group as the third tomographic image group.
The processor
The image processing apparatus according to claim 1 or 2, wherein the composite two-dimensional image in which the density of the second region is controlled to a density close to that of the third region is generated.
In the third tomographic image group, the absolute difference between the pixel value of the attention pixel among the plurality of tomographic images and the average value of the pixel values of the attention pixels of all the plurality of tomographic images is absolute. A tomographic image group whose value is equal to or higher than a preset threshold, a tomographic image group having a preset number of tomographic images in descending order of the dispersion value of the pixel value of the region of interest including the pixel of interest among the plurality of tomographic images, or the plurality of faults. The image processing apparatus according to any one of claims 1 to 5, which is a tomographic image group having pixels whose edges are detected by edge detection processing in an image.
The processor
Claimed from claim 1 which controls to display the generated synthetic two-dimensional image and controls to display the determination result of whether it is a mass or a local mammary gland mass on the synthetic two-dimensional image. Item 6. The image processing apparatus according to any one of Items 6.
Tumor candidate areas are detected from multiple tomographic images representing multiple tomographic planes of the subject.
For each of the detected tumor candidate areas, it is determined whether it is a tumor or a local mass of the mammary gland.
In the first region determined to be a tumor, the first tomographic image group is selected from the plurality of tomographic images.
In the second region determined to be a local mammary gland mass, a second tomographic image group is selected from the plurality of tomographic images.
In the third region other than the first region and the second region, a third tomographic image group is selected from the plurality of tomographic images.
An image processing method in which a processor included in an image processing apparatus executes a process of generating a composite two-dimensional image using a tomographic image group selected for each of the first region, the second region, and the third region.
Tumor candidate areas are detected from multiple tomographic images representing multiple tomographic planes of the subject.
For each of the detected tumor candidate areas, it is determined whether it is a tumor or a local mass of the mammary gland.
In the first region determined to be a tumor, the first tomographic image group is selected from the plurality of tomographic images.
In the second region determined to be a local mammary gland mass, a second tomographic image group is selected from the plurality of tomographic images.
In the third region other than the first region and the second region, a third tomographic image group is selected from the plurality of tomographic images.
Image processing for causing a processor included in the image processing apparatus to perform processing for generating a composite two-dimensional image using a tomographic image group selected for each of the first region, the second region, and the third region. program.